Medical errors causing preventable adverse events are the third leading cause of death in the U.S. The role of human cognition in contributing to errors in complex healthcare environments is increasingly being recognized. Many studies have demonstrated a direct relationship between cognitive performance metrics and patient outcomes. Notwithstanding, the role of cognition in improving safety has not received enough attention. Cardiac surgical OR is a high-impact area to improve patient safety and save lives. High mental workload during complex cardiac surgery and associated reduction of spare attentional capacity is bound to have a detrimental effect on a team's ability to detect their own errors. In the proposed project, we will explore and evaluate novel approaches to using process and patient real-time data, and device status information dynamically and proactively to provide guidance forward from the current surgical state. The proposed Cognition-based Guidance System will be deployed intraoperatively. We believe that human errors in the cardiac OR can be reduced by supporting individual and team interactions via dynamically-updated electronic guidance that are responsive to the details of the on-going process (i.e., context-aware), to the patient's current medical condition (i.e., patient-aware), and to cognitive workload demands on the participating clinicians (i.e., workload-aware).